search

ARS-SWRC / agwa

Toolbox and code (ArcPy/Python) for the Automated Geospatial Watershed Assessment tool (AGWA).
Other
6 stars 1 forks source link

Failed to Execute (ParametrizeLandCoverandSoils) #6

Open kennethelsner opened 3 months ago

kennethelsner commented 3 months ago

Good Afternoon,

I have successfully run AGWA in the State of Oregon. I am running AGWA in a new part of Oregon and I am receiving the following error during Parametrization. The same gSSURGO data and LC data worked in another part of the state. Is texture fragmental material in the AGWA lookup table the issue? If yes, is there a way to tweak the table to allow parameterization to complete successfully?

image

Thanks!

haiyan-Wei commented 2 months ago

Thank you for bringing this to my attention. That's right, Fragmental is not in the lookup table yet. I’ll look into it as soon as possible. If you’re familiar with this texture type and comfortable tweaking the table on your end, you can edit the table kin_lut in the lookup_tables.gdb file, located in the AGWApro_v0.5_directory folder.

haiyan-Wei commented 2 months ago

@kennethelsner The "Fragmental material" texture is now available in the AGWA lookup table. We've also updated the Ks and G values for the most common soil texture types based on our latest knowledge. I also fixed a bug in Step 5.

The new code is ready to go, please test it with the updated "AGWApro V0.5 Directory" available here: https://www.tucson.ars.ag.gov/agwa/downloads/.

I appreciate your patience.

kennethelsner commented 2 months ago

At least I got a new error this time? image

jburgessconforti commented 2 months ago

step4agwa

Since we've started this thread I've gone from having it break on step 5 to having it break on step 4...

haiyan-Wei commented 2 months ago

At least I got a new error this time? image

@kennethelsner It looks like AGWA was looking for table component in folder agwa_gwen, which does not look like a file geodatabase. I'm curious what's your input for step 19.6? If you uncheck 19.6, an input box should show up and let you choose the file geodatabase where table component is located. Could you check if that's the right directory for table component?

haiyan-Wei commented 2 months ago

@jburgessconforti Thanks for sharing the error message. I think I know what has caused this error. I'm on it!

jburgessconforti commented 2 months ago

@haiyan-Wei So I was able to get rid of the none of dem.tif was found error message but shoving everything in to a .gdb. It looks like arc or the code wasn't working well with the periods in the file extension. I'm getting an error still on step 4 but its only about floats and none types (see below). I appreciate the quick response and all the work you're doing on this.

step4error2

haiyan-Wei commented 2 months ago

@jburgessconforti Thanks for letting me know. I'm glad you got it! This has happened to me before. I thought I had made sure the issue with the file extension was taken care of. I'll double-check!

kennethelsner commented 2 months ago

@haiyan-Wei The soil database was totally my fault. Thanks for pointing it out.

I was picking up a project that I hadn't touched in few weeks and I didn't have my default soil database in the correct location. Pointing to the gSSURGO database fixed my issue in Step 5. I made it all the way to Executing the Sim which gave me the following conflicting error. CMD prompt says error while Pro says successful. That type of conflicting messaging is confusing for users. image

haiyan-Wei commented 2 months ago

@kennethelsner

I believe AGWA sets a default value for the slope of very flat channels as a small number to prevent this error. If not, we need to change that.

I wonder if something went wrong in Step 4. Could you open the parameters_channels table located in the file geodatabase for your delineation and check if the other channel parameters look reasonable?

For example, in the case of the tutorial, it would be the parameters_channels table in E:\agwa_prj\learn_agwa\fire_demo\AndreasCanyon\AndreasCanyon.gdb.

Thanks!

image

kennethelsner commented 2 months ago

@haiyan-Wei I think it is a great idea to give very flat channels a small number to prevent the error.

It does not appear, that agwa is honoring any default value. 2 negative values in meanslope

image

haiyan-Wei commented 2 months ago

@kennethelsner Thank you for your feedback. I've made the changes, along with a few minor adjustments to the interfaces of several steps. The channel slope is calculated based on upstream and downstream elevations, which could sometimes result in a negative value for flat channels. I’ve added a minimal threshold of 0.0001.

If you'd like to test your watershed again, please let me know if everything works as expected this time. Thanks again!

haiyan-Wei commented 2 months ago

Hello Capina,

Thank you for the message, from another message threshold, it looked like the new code worked out for you and this no longer is an issue. Let me know if it is otherwise.

Best, Haiyan

On Fri, Sep 6, 2024 at 11:13 PM Captina @.***> wrote:

Well I finally made it past step but I did have to add two more textures, Highly and Moderately decomposed plant material to the Kin_lut table.

Now its failing at step five with this message:

Traceback (most recent call last): File "C:\AGWA\agwa-pro-toolbox_v0.5_beta\code\src\tool_parameterize_land_cover_and_soils.py", line 364, in execute agwa.parameterize(prjgdb, workspace, delineation, discretization, parameterization_name, save_intermediate_outputs) File "C:\AGWA\agwa-pro-toolbox_v0.5_beta\code\src\code_parameterize_land_cover_and_soils.py", line 68, in parameterize parameterize_hillslopes(workspace, delineation_name, discretization_name, parameterization_name, File "C:\AGWA\agwa-pro-toolbox_v0.5_beta\code\src\code_parameterize_land_cover_and_soils.py", line 91, in parameterize_hillslopes df_cover = intersect_weight_land_cover_by_area(workspace, delineation_name, discretization_name, land_cover, File "C:\AGWA\agwa-pro-toolbox_v0.5_beta\code\src\code_parameterize_land_cover_and_soils.py", line 217, in intersect_weight_land_cover_by_area apply_buffer_flag, buffer_size = is_raster_larger_and_buffer(land_cover, watershed_feature_class) File "C:\AGWA\agwa-pro-toolbox_v0.5_beta\code\src\code_parameterize_land_cover_and_soils.py", line 849, in is_raster_larger_and_buffer raise Exception(f"The raster {raster} does not cover the entire watershed. " Exception: The raster E:\ArcGIS\Layers\nlcd_2019_land_cover_l48_20210604.img does not cover the entire watershed. Please provide a raster that covers the entire watershed. Failed to execute (ParameterizeLandCoverAndSoils).

Despite the NLCD raster being well large enough to cover the watershed

— Reply to this email directly, view it on GitHub https://github.com/ARS-SWRC/agwa/issues/6#issuecomment-2335084557, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMT4ZZ2IDN4EPZHF22JYYY3ZVKKQPAVCNFSM6AAAAABMSXGSUKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMZVGA4DINJVG4 . You are receiving this because you were mentioned.Message ID: @.***>